Search

In some areas of Antarctica, blue‐colored bare ice is exposed at the surface. These blue ice areas (BIAs) can trap meteorites or old ice and are vital for understanding the climatic history. By combining multi‐sensor remote sensing data (MODIS, RADARSAT‐2, and TanDEM‐X) in a deep learning framework, we map blue ice across the continent at 200‐m resolution. We use a novel methodology for image segmentation with "noisy" labels to learn an underlying "clean" pattern with a neural network. In total, BIAs cover ca. 140,000 km2 (∼1%) of Antarctica, of which nearly 50% located within 20 km of the grounding line. There, the low albedo of blue ice enhances melt‐water production and its mapping is crucial for mass balance studies that determine the stability of the ice sheet. Moreover, the map provides input for fieldwork missions and can act as constraint for other geophysical mapping efforts. Plain Language Summary: While most of the continent of Antarctica is covered by snow, in some areas, ice is exposed at the surface, with a typical blue color. At lower elevations, blue ice enhances melt‐water production, which is important for studying the future of the ice sheet. Moreover, scientific teams frequently visit blue ice areas (BIAs) as they act as traps for meteorites and very old ice. In this study, we map the extent and the exact location of BIAs using various satellite observations. These diverse observations are efficiently combined in an artificial intelligence algorithm. We develop the algorithm so that it can learn to map blue ice even though existing training labels, which teach the algorithm what blue ice looks like, are imperfect. We quantify that the new map scores better on various performance metrics compared to the current most‐used blue ice map. Moreover, for the first time, we estimate uncertainties of the detection of blue ice. The map indicates that ca. 1% of the surface of Antarctica exposes blue ice and will be important for fieldwork missions and understanding surface processes leading to melt and potential sea level rise. Key Points: We map blue ice areas in Antarctica by combining multi‐sensor satellite observations in a convolutional neural networkBlue ice covers ca. 140,000 km2 (∼1%) of Antarctica, of which ca. 50% located in the grounding zoneOur map will improve mass balance estimates and studies on ice‐shelf stability, and will support searches for meteorites or old ice [ABSTRACT FROM AUTHOR]

Accurately estimating the surface melt volume of the Antarctic Ice Sheet is challenging and has hitherto relied on climate modeling or observations from satellite remote sensing. Each of these methods has its limitations, especially in regions with high surface melt. This study aims to demonstrate the potential of improving surface melt simulations with a regional climate model by deploying a deep learning model. A deep-learning-based framework has been developed to correct surface melt from the regional atmospheric climate model version 2.3p2 (RACMO2), using meteorological observations from automatic weather stations (AWSs) and surface albedo from satellite imagery. The framework includes three steps: (1) training a deep multilayer perceptron (MLP) model using AWS observations, (2) correcting Moderate Resolution Imaging Spectroradiometer (MODIS) albedo observations, and (3) using these two to correct the RACMO2 surface melt simulations. Using observations from three AWSs at the Larsen B and C ice shelves, Antarctica, cross-validation shows a high accuracy (root-mean-square error of 0.95 mm w.e. d -1 , mean absolute error of 0.42 mm w.e. d -1 , and a coefficient of determination of 0.95). Moreover, the deep MLP model outperforms conventional machine learning models and a shallow MLP model. When applying the trained deep MLP model over the entire Larsen Ice Shelf, the resulting corrected RACMO2 surface melt shows a better correlation with the AWS observations for two out of three AWSs. However, for one location (AWS 18), the deep MLP model does not show improved agreement with AWS observations; this is likely because surface melt is largely driven by factors (e.g., air temperature, topography, katabatic wind) other than albedo within the corresponding coarse-resolution model pixels. Our study demonstrates the opportunity to improve surface melt simulations using deep learning combined with satellite albedo observations. However, more work is required to refine the method, especially for complicated and heterogeneous terrains. [ABSTRACT FROM AUTHOR]

Areas of structural damage mechanically weaken Antarctic ice shelves. This potentially preconditions ice shelves for disintegration and enhanced grounding line retreat. The development of damage and its feedback on marine ice sheet dynamics has been identified as key to future ice shelf stability and sea level contributions from Antarctica. However, it is one of the least understood processes that impact ice shelf instability since quantifying damage efficiently and accurately is a challenging task. Challenges relate to the complex surface of Antarctica, variations in viewing-illumination geometry, snow or cloud cover and variable signal-to-noise levels in satellite imagery. Therefore, automated damage assessment approaches require careful pre- and post-processing, lacking the option to be applied to wider spatiotemporal domains. Simultaneously, studies that use manual mapping are usually limited due to the effort required for extensive mapping, which either results in a limited spatial domain or the use of low resolution data. This study proposes the NormalisEd Radon transform Damage detection (NeRD) method to detect damage features and their orientations from multi-source satellite imagery. NeRD performs robust, high resolution, large-scale damage assessments. NeRD is applied to the ice shelves in the Amundsen Sea Embayment (ASE) and validated with both manually labelled and existing fracture maps. Validation shows that NeRD detects damage with high recall and provides an accurate physical representation of multi-scale damage features and their orientation. Sensitivity analyses indicate NeRD is robust to different resolution parameter settings. NeRD consistently detects damage for different data sources ranging from optical Landsat 7/8 and Sentinel-2 optical to Synthetic Aperture Radar Sentinel-1 data. Therefore, NeRD paves the way for synergistic multi-source damage detection that overcomes remaining limitations from individual sources. Results show varying damage patterns on the ice shelves in the ASE area in austral summer 2020–2021, with most damage located on the Pine Island, Crosson and Thwaites ice shelves. We show a damage increase on the Pine Island ice shelf between 2013–2019, and display advection and rotation of crevasses. The detected damage orientation can provide insight in the type of crevasse opening mode and the development of damage over time. The damage maps produced with NeRD can help evaluate ice sheet models or machine learning approaches, improving our understanding of damage evolution. • We present The NormalisEd Radon transform Damage detection (NeRD) method. • The NeRD method consistently detects damage across spatiotemporal domains. • The NeRD method detects multi-scale damage features and their orientations. • The NeRD method is directly applicable on multi-source satellite data. [Display omitted] [ABSTRACT FROM AUTHOR]

The role of clouds on the surface mass balance of ice sheets remains a point of debate as there is no scientific consensus on the role of the cloud radiative effect (CRE) on the two major ice sheets. For the Greenland Ice Sheet, for example, conflicting studies argue either that clouds reduce surface melt through their radiative effect by blocking short-wave radiation (cloud cooling) or enhance surface melt by reducing meltwater refreezing (cloud warming). For the Antarctic Ice Sheet, on the other hand, the impact of the CRE effects on the ice sheet remain poorly quantified, although it can play a major role on the melt processes on ice shelves.To assess the impact of the CRE on surface snow and firn conditions on both ice sheets, simulations have been performed with SNOWPACK, a physically-based snow model forced with and without cloud radiative forcing. For this purpose, a state-of-the-art hybrid dataset has been created to represent the cloud forcing on the surface mass balance. This hybrid data set is a combination of CloudSat-CALIPSO satellite observations and data from the regional climate model, RACMO2.3. The combination of both data sets results in an observation-based forcing dataset with higher temporal resolution than the satellite observations can provide. Simulations are performed such that the response of the firn can be separated in a short-term and long-term component.Results show that the seasonal variation of the CRE is positive throughout the year, which implies a net cloud warming. Additionally, the simulations highlight that cloud cooling and cloud warming occur on different time scales: the first being a more direct, short-term effect and the latter having a long-term effect by initiating a feedback response that changes firn conditions. The long-term response of the firn to the CRE is shown to be dominant in summer, enhancing meltwater production and runoff by initializing a melt-albedo feedback.The long-term CRE impact on firn conditions during the melt season highlights the importance of the (initial) firn conditions when performing cloud radiation studies. It calls for the need of either detailed information of the firn surface layers or the inclusion of a snow-model that can accurately model firn response in such studies. [ABSTRACT FROM AUTHOR]

While the influence of surface melt on Antarctic ice shelf stability can be large, the duration and affected area of melt events are often small. Therefore, melt events are difficult to capture with remote sensing, as satellite sensors always face the trade-off between spatial and temporal resolution. To overcome this limitation, we developed UMelt: a surface melt record for all Antarctic ice shelves with a high spatial (500 m) and high temporal (12 h) resolution for the period 2016–2021. Our approach is based on a deep learning model, specifically a U-Net, which was developed in Google Earth Engine. The U-Net combines microwave remote sensing observations from three sources: Sentinel-1, Special Sensor Microwave Imager/Sounder (SSMIS), and Advanced Scatterometer (ASCAT). The U-Net was trained on the Shackleton Ice Shelf for melt seasons 2017–2021, using the fine-scale melt patterns of Sentinel-1 as reference data and SSMIS, ASCAT, a digital elevation model, and multi-year Sentinel-1 melt fraction as predictors. The trained U-Net performed well on the Shackelton Ice Shelf for test melt season 2016–2017 (accuracy: 91.3%; F1-score: 86.9%), and the Larsen C Ice Shelf, which was not considered during training (accuracy: 91.0%; F1-score: 89.3%). Using the trained U-Net model, we have successfully developed the UMelt record. UMelt allows Antarctic-wide surface melt to be detected at a small scale while preserving a high temporal resolution, which could lead to new insights into the response of ice shelves to a changing atmospheric forcing. • Satellite resolution trade-offs hamper accurate surface melt detection in Antarctica. • We present UMelt: An Antarctic-wide, high-resolution surface melt record. • UMelt was developed by merging multi-source remote sensing data using deep learning. • UMelt detects intricate surface melt events that were difficult to capture before. [ABSTRACT FROM AUTHOR]